ArcFace: Additive Angular Margin Loss for Deep Face Recognition
Citations
6 citations
Cites background from "ArcFace: Additive Angular Margin Lo..."
...With the squeeze-and-excitation optimization[7], [4, 48, 6] achieves slightly better performance on face recognition and image classification tasks....
[...]
...35% computational cost compared to large-scale deep face recognition network[4]....
[...]
6 citations
6 citations
6 citations
6 citations
Cites methods from "ArcFace: Additive Angular Margin Lo..."
...The optimization of the parameters is done by minimizing a weighted sum of three losses: triplet loss (Weinberger, Saul, 2009), ArcFace loss (Deng et al., 2019), and focal loss (Lin et al., 2017)....
[...]
...The optimization of the parameters is done by minimizing a weighted sum of three losses: triplet loss (Weinberger, Saul, 2009), ArcFace loss (Deng et al., 2019), and focal loss (Lin et al....
[...]
References
123,388 citations
33,597 citations
30,843 citations
13,268 citations
10,447 citations